SUMMARY Alzheimer's Disease (AD) is thought to involve early synapse loss. Familial AD is most often caused by mutations in presenilins, which are the catalytic subunits of g-secretase. Inactivation of g-secretase in adult mice by conditional deletion of its presenilin or nicastrin subunits causes synaptic impairments followed by neurodegeneration, but the relation of the loss of g-secretase acitivity, the observed synaptic impairments, and the neurodegeneration is unclear, as is the connection between these processes and human AD. The present project will focus on clarifying the synaptic function of g-secretase in mature neurons and its relation to neurodegeneration, using mice as a model system. The project proposes four specific aims to address this overall goal. The first two specific aims will mechanistically characterize the synaptic impairments that are caused by g-secretase inactivation in young adult mice in vivo, and analyze their relation to the neurodegeneration that develops at a later stage after g-secretase inactivation. The third and fourth specific aim will then test the hypothesis that at least some of the synaptic functions of g-secretase that are impaired upon its inactivation may be mediated by g-secretase-dependent cleavage of presynaptic neurexins and postsynaptic neuroligins, which are trans-synaptic cell-adhesion molecules that bind to each other and act as master regulators of synaptic properties. Strikingly in this context, neurexins and neuroligins were shown previously to be substrates for g-secretase, were genetically linked to AD, and are arguably the most plausible mediators of g-secretase function at the synapse. To test the involvement of neurexins and neuroligins in the synaptic functions of g-secretase, the project will characterize the site and regulation of the g-secretase- dependent cleavage of neurexins and neuroligins, and probe the function of this cleavage using conditional knockout mice of these molecules. Moreover, the project will examine whether inactivation of neurexin- and/or neuroligin-cleavage promotes neurodegeneration. Viewed together, the experiments of this project will thus not only characterize the synaptic function of g-secretase and its relation to neurodegeneration, but also determine whether the synaptic function of g-secretase involves the cleavage of neurexins and/or neuroligins and whether such cleavage may play a contributory role in the pathogenesis of AD.